Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    polite

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     paradox
    -0.07
     causal
    -0.07
    _inc
    -0.07
    _assignment
    -0.07
     challenge
    -0.07
     Challenge
    -0.07
    graded
    -0.07
     hardware
    -0.07
     challenges
    -0.06
     Sponsor
    -0.06
    POSITIVE LOGITS
     polite
    0.17
     politely
    0.14
     polit
    0.07
     courteous
    0.07
     diplomatic
    0.07
     İ
    0.06
    ��
    0.06
     Phones
    0.06
    lara
    0.06
     diye
    0.06
    Act Density 0.004%

    No Known Activations